bugfix: remove out large samples from the multi pack batch sampler #52
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ChanderG
force-pushed
the
bugfix-large-datapoints
branch
from
086f887 to
f8ae22b
Compare
|
Great catch! This should be processed in the data processing step. Can you add a warning in red or yellow 033 notation so the user knows? |
|
You can just add it when there's a sample larger than Max batch lenght |
larger entries, longer than max_batch_len, may be present in the output of the sampler which gets filtered out in the collator in rare instances, this can lead to empty batches which can be problematic for the training loop Signed-off-by: ChanderG <[email protected]>
ChanderG
force-pushed
the
bugfix-large-datapoints
branch
from
f8ae22b to
72fde51
Compare
aldopareja
approved these changes
Jun 24, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fix for #51
larger entries, longer than max_batch_len, may be present in the output of the sampler which gets filtered out in the collator
in rare instances, this can lead to empty batches which can be problematic for the training loop